summaryrefslogtreecommitdiffstats
path: root/libutils
diff options
context:
space:
mode:
Diffstat (limited to 'libutils')
-rw-r--r--libutils/String8.cpp25
-rw-r--r--libutils/Unicode.cpp15
-rw-r--r--libutils/tests/String8_test.cpp19
3 files changed, 43 insertions, 16 deletions
diff --git a/libutils/String8.cpp b/libutils/String8.cpp
index ad65fdb..75dfa29 100644
--- a/libutils/String8.cpp
+++ b/libutils/String8.cpp
@@ -102,20 +102,21 @@ static char* allocFromUTF16(const char16_t* in, size_t len)
{
if (len == 0) return getEmptyString();
- const ssize_t bytes = utf16_to_utf8_length(in, len);
- if (bytes < 0) {
+ // Allow for closing '\0'
+ const ssize_t resultStrLen = utf16_to_utf8_length(in, len) + 1;
+ if (resultStrLen < 1) {
return getEmptyString();
}
- SharedBuffer* buf = SharedBuffer::alloc(bytes+1);
+ SharedBuffer* buf = SharedBuffer::alloc(resultStrLen);
ALOG_ASSERT(buf, "Unable to allocate shared buffer");
if (!buf) {
return getEmptyString();
}
- char* str = (char*)buf->data();
- utf16_to_utf8(in, len, str);
- return str;
+ char* resultStr = (char*)buf->data();
+ utf16_to_utf8(in, len, resultStr, resultStrLen);
+ return resultStr;
}
static char* allocFromUTF32(const char32_t* in, size_t len)
@@ -124,21 +125,21 @@ static char* allocFromUTF32(const char32_t* in, size_t len)
return getEmptyString();
}
- const ssize_t bytes = utf32_to_utf8_length(in, len);
- if (bytes < 0) {
+ const ssize_t resultStrLen = utf32_to_utf8_length(in, len) + 1;
+ if (resultStrLen < 1) {
return getEmptyString();
}
- SharedBuffer* buf = SharedBuffer::alloc(bytes+1);
+ SharedBuffer* buf = SharedBuffer::alloc(resultStrLen);
ALOG_ASSERT(buf, "Unable to allocate shared buffer");
if (!buf) {
return getEmptyString();
}
- char* str = (char*) buf->data();
- utf32_to_utf8(in, len, str);
+ char* resultStr = (char*) buf->data();
+ utf32_to_utf8(in, len, resultStr, resultStrLen);
- return str;
+ return resultStr;
}
// ---------------------------------------------------------------------------
diff --git a/libutils/Unicode.cpp b/libutils/Unicode.cpp
index fb876c9..2b5293e 100644
--- a/libutils/Unicode.cpp
+++ b/libutils/Unicode.cpp
@@ -14,6 +14,7 @@
* limitations under the License.
*/
+#include <log/log.h>
#include <utils/Unicode.h>
#include <stddef.h>
@@ -182,7 +183,7 @@ ssize_t utf32_to_utf8_length(const char32_t *src, size_t src_len)
return ret;
}
-void utf32_to_utf8(const char32_t* src, size_t src_len, char* dst)
+void utf32_to_utf8(const char32_t* src, size_t src_len, char* dst, size_t dst_len)
{
if (src == NULL || src_len == 0 || dst == NULL) {
return;
@@ -193,9 +194,12 @@ void utf32_to_utf8(const char32_t* src, size_t src_len, char* dst)
char *cur = dst;
while (cur_utf32 < end_utf32) {
size_t len = utf32_codepoint_utf8_length(*cur_utf32);
+ LOG_ALWAYS_FATAL_IF(dst_len < len, "%zu < %zu", dst_len, len);
utf32_codepoint_to_utf8((uint8_t *)cur, *cur_utf32++, len);
cur += len;
+ dst_len -= len;
}
+ LOG_ALWAYS_FATAL_IF(dst_len < 1, "dst_len < 1: %zu < 1", dst_len);
*cur = '\0';
}
@@ -324,7 +328,7 @@ int strzcmp16_h_n(const char16_t *s1H, size_t n1, const char16_t *s2N, size_t n2
: 0);
}
-void utf16_to_utf8(const char16_t* src, size_t src_len, char* dst)
+void utf16_to_utf8(const char16_t* src, size_t src_len, char* dst, size_t dst_len)
{
if (src == NULL || src_len == 0 || dst == NULL) {
return;
@@ -345,9 +349,12 @@ void utf16_to_utf8(const char16_t* src, size_t src_len, char* dst)
utf32 = (char32_t) *cur_utf16++;
}
const size_t len = utf32_codepoint_utf8_length(utf32);
+ LOG_ALWAYS_FATAL_IF(dst_len < len, "%zu < %zu", dst_len, len);
utf32_codepoint_to_utf8((uint8_t*)cur, utf32, len);
cur += len;
+ dst_len -= len;
}
+ LOG_ALWAYS_FATAL_IF(dst_len < 1, "%zu < 1", dst_len);
*cur = '\0';
}
@@ -408,10 +415,10 @@ ssize_t utf16_to_utf8_length(const char16_t *src, size_t src_len)
const char16_t* const end = src + src_len;
while (src < end) {
if ((*src & 0xFC00) == 0xD800 && (src + 1) < end
- && (*++src & 0xFC00) == 0xDC00) {
+ && (*(src + 1) & 0xFC00) == 0xDC00) {
// surrogate pairs are always 4 bytes.
ret += 4;
- src++;
+ src += 2;
} else {
ret += utf32_codepoint_utf8_length((char32_t) *src++);
}
diff --git a/libutils/tests/String8_test.cpp b/libutils/tests/String8_test.cpp
index c42c68d..7cd67d3 100644
--- a/libutils/tests/String8_test.cpp
+++ b/libutils/tests/String8_test.cpp
@@ -17,6 +17,7 @@
#define LOG_TAG "String8_test"
#include <utils/Log.h>
#include <utils/String8.h>
+#include <utils/String16.h>
#include <gtest/gtest.h>
@@ -72,4 +73,22 @@ TEST_F(String8Test, OperatorPlusEquals) {
EXPECT_STREQ(src3, " Verify me.");
}
+// http://b/29250543
+TEST_F(String8Test, CorrectInvalidSurrogate) {
+ // d841d8 is an invalid start for a surrogate pair. Make sure this is handled by ignoring the
+ // first character in the pair and handling the rest correctly.
+ String16 string16(u"\xd841\xd841\xdc41\x0000");
+ String8 string8(string16);
+
+ EXPECT_EQ(4U, string8.length());
+}
+
+TEST_F(String8Test, CheckUtf32Conversion) {
+ // Since bound checks were added, check the conversion can be done without fatal errors.
+ // The utf8 lengths of these are chars are 1 + 2 + 3 + 4 = 10.
+ const char32_t string32[] = U"\x0000007f\x000007ff\x0000911\x0010fffe";
+ String8 string8(string32);
+ EXPECT_EQ(10U, string8.length());
+}
+
}